Funded Research Projects

ProbInG: Distribution Recovery for Invariant Generation of Probabilistic Programs (WWTF - Vienna Science and Technology Fund), opens an external URL in a new window

Principal Investigators: Ezio Bartocci, opens an external URL in a new window (PI), Laura Kovács, opens an external URL in a new window (Co-PI), Efstathia Bura, opens an external URL (Co-PI)

Starting Date: May 2020 

Abstract

Probabilistic programming is a new emerging paradigm adopted by high-tech giants, such as Google, Amazon and Uber, to simplify the development of AI/machine learning based applications, such as route planning and detecting cyber intrusions. Probabilistic programming languages include native constructs for sampling distributions allowing to freely mix deterministic and stochastic elements. The resulting flexible framework comes at the price of programs with behaviors hard to analyze, leading to unpredictable adverse consequences in safety-critical applications. One of the main challenges in the analysis of these programs is to compute invariant properties that summarize loop behaviors. Despite recent results, full automation of invariant generation is at its infancy and only targets expected values of the program variables, which is insufficient to recover the full probabilistic program behavior. Our project aims at developing novel and fully automated approaches to generate invariants over higher-order moments and the value distribution of program variables, without any user guidance. We will employ methods from symbolic summation, polynomial algebra and statistics and combine them with static analysis techniques. Our results will reduce the need of expert knowledge in ensuring the safety and security of computer systems and will cut the design costs of applications based on probabilistic programs, bringing crucial intellectual and economic benefits to our society.

SecInt Doctoral College: Statistical Verification of Security Properties for Cyber- Physical Systems., opens an external URL in a new window

Starting Date: July 2020

Previous projects:

Sufficient Dimension Reduction Methodology in Forecasting (FWF - Austrian Science Fund), opens an external URL in a new window

Project leader: Efstathia Bura, opens an external URL
Project Assistants: Daniel Kapla, opens an external URL, Andrey Kofnov, opens an external URL in a new window
Starting Date: December 2017

Former Project Assistants: Karl Oskar Ekvall, Lukas Fertl, Barbara Brune,

Abstract
Economists and policy makers have more data at their disposal than ever before. Extracting the most relevant information prevents reacting to idiosyncratic movements and can lead to more precise forecasts and macro/microeconomic analyses. However, how to use these data effectively is an open problem.

Dynamic Factor Models are pervasive in macro-econometrics and financial econometrics for both measuring co-movement and forecasting time series. However, the data reduction in DFMs comprises of summarizing the information in large data sets with a few components that capture a large proportion of their total variability without considering the forecasting ability of the reduced data.

Sufficient Dimension Reduction (SDR) is a collection of tools for reducing the dimension of multivariate data in regression problems without losing inferential information for modeling the response. SDR uses many noisy signals in the observable data to extract information about the underlying structural sources of comovement that can be used to inform the building of forecasting models.

This project will extend existing and develop new SDR methodology in econometric modeling and forecasting. SDR methods and data analysis tools will be developed to identify and estimate exhaustive reductions, including nonlinear data reductions, which have been marginally investigated in the DFM context.

Furthermore, SDR methods for (a) targeted PCA and (b) for large p-small T (many predictors, few observations) time series regressions based on Krylov subspaces will be developed.

Envelope models for multivariate response forecasting, such as central banks' macro forecasts, will be also developed and applied.

The proposed research intends to make a significant contribution to the development of statistical tools that reduce data complexity in order to understand and model the underlying relationships and structures that drive the economy and obtain more accurate forecasts.

TU Wien Research Database, opens an external URL in a new window